Cancer Therapy: Preclinical

The Targeted Immunocytokine L19-IL2 Efficiently Inhibits the

Growth of Orthotopic Pancreatic Cancer
Karola Wagner,1 Petra Schulz,1 Arne Scholz,1 Bertram Wiedenmann,1 and Andreas Menrad2

Abstract Purpose: Effective control of pancreatic cancer has been hampered primarily by the lack of tumor
specificity of current treatment modalities. The highly specific antibody-mediated delivery of
therapeutic agents to the tumor microenvironment might overcome this problem. We therefore
investigated the therapeutic efficacy of the targeted immunocytokine L19-Interleukin-2 (L19-IL2),
consisting of the human single-chain Fv antibody L19, which is highly specific for the extradomain
B (ED-B) of fibronectin, and the human cytokine IL-2, in pancreatic cancer.
Experimental Design: Therapeutic effects of L19-IL-2, IL-2, and gemcitabine on tumor growth
and metastasis were evaluated in orthotopic mouse models for pancreatic cancer. Immunohisto-
chemistry was done to define ED-B expression, tumor necrosis, apoptosis, proliferation, and
invasion of macrophages and natural killer (NK) cells. NK cells were depleted by i.v. injection of
an anti-asialo-GM-1antibody.
Results: ED-B is selectively expressed in human pancreatic cancer and in primary tumors and
metastases of the mouse models. L19-IL-2 therapy was clearly superior to untargeted IL-2 or
gemcitabine and inhibited tumor growth and metastasis with remarkable long-term tumor control.
Therapeutic effects were associated with the induction of extensive tumor necrosis and inhibition
of tumor cell proliferation. Immunohistochemistry revealed an increase of macrophages and NK
cells in the tumor tissue, suggesting immune-mediated mechanisms. The functional relevance of
NK cells for the therapeutic effect of the targeted immunocytokine L19-IL-2 was confirmed by NK
cell depletion, which completely abolished its antitumor efficacy.
Conclusions: These preclinical results strongly encourage the initiation of clinical studies using
L19-IL-2 in pancreatic cancer.

Adenocarcinoma of the pancreas represents the fifth leading
cause of cancer related death in industrialized western countries
(1). The prognosis of patients diagnosed with pancreatic cancer
is extremely poor with an estimated overall 5-year survival rate
of 1% to 4%. Surgical resection provides the only potentially
curative treatment, but locally extended or metastasized disease
precludes surgical treatment in most cases (2, 3). Due to poor
selectivity, toxicities against nonmalignant cells, and dose
limitations, currently available strategies including palliative,
Authors’ Affiliations: 1Department of Hepatology and Gastroenterology, Charite¤,
CampusVirchow-Klinikum, Humboldt-University; and 2Anti-Angiogenesis Research,
Corporate Research Oncology, Bayer-Schering Pharma AG, Berlin, Germany
Received 1/20/08; revised 3/6/08; accepted 3/10/08.
Grant support: Deutsche Forschungsgemeinschaft grant Scho-848/1-1,1-2
(A. Scholz).
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: K.Wagner and P. Schulz contributed equally to this publication.
Current address forA. Menrad: Genzyme Europe Research, 310 Cambridge Science
Park, Cambridge CB4 0WG, United Kingdom.
Dedicated to Prof. Stefan Rosewicz (1960-2004) who contributed significantly to
this work.
Requests for reprints: Petra Schulz, Depar tment of Hepatology and
Gastroenterology, Charite¤, Campus Virchow-Klinikum, Humboldt-University,
Augustenburger Platz 1, 13353 Berlin, Germany. Phone: 49-30-450-559-708;
Fax: 49-30-450-559-939; E-mail: Petra.Schulz@ charite.de.
F 2008 American Association for Cancer Research.
doi:10.1158/1078-0432.CCR-08-0157
chemotherapeutic, and radiation treatments have little effect on
the aggressive course of disease (4 – 7). One promising avenue
to overcome these obstacles is the highly specific antibody-
mediated delivery of therapeutic agents to the tumor microen-
vironment. In particular, vascular and/or stromal targeting
represents an appealing therapeutic strategy for several reasons.
First, targets, which are selectively expressed around tumor
vessels and/or in the tumor stroma, are easily accessible to
systemically administered antibodies. Second, markers of
tumor stroma and tumor neovasculature are typically produced
by endothelial cells and/or myofibroblasts, which are geneti-
cally more stable than tumor cells. Third, as neoangiogenesis is
a prerequisite of tumor growth and metastasis, the selective
targeting approach of L19-Interleukin-2 (IL-2) to newly
forming tumor blood vessels should result in a therapeutical
benefit. Currently, one of the most selective oncofetal antigens
associated with neoangiogenesis and tumor growth is the
extradomain B (ED-B) of fibronectin (6, 8). The fibronectin
splice variant ED-B, a small domain of 91 amino acids, which is
homologous from mouse to man, is usually absent in both
plasma and tissue-fibronectin, except for some blood vessels of
the regenerating endometrium and the ovaries (6, 8). However,
it may become inserted in the fibronectin molecule during
active tissue remodeling associated with neoangiogenesis,
thereby accumulating around the neovasculature and in the
stroma of malignant tumors and other tissues undergoing
angiogenesis (6, 8). Recently, a human single-chain Fv (scFv)
antibody fragment L19 has been generated, which displays

www.aacrjournals.org 4951 Clin Cancer Res 2008;14(15) August 1, 2008

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
Cancer Therapy: Preclinical
picomolar binding affinity for ED-B, and has been verified to
selectively target tumor neovasculature in both experimental
tumor models (9) and patients with cancer (10), thus paving
the way for a novel therapeutic approach targeting tumor
neovasculature.
IL-2 has been characterized as one of the most potent
antitumor cytokines. However, despite being approved for the
clinical treatment of metastatic renal cell carcinoma, systemi-
cally applied IL-2 has failed to fulfill earlier hopes. In part, this
is due to serious, potentially life-threatening side effects that
limit dose escalation and prevent the application of sufficiently
high doses (11, 12). In addition, the fast clearance of
systemically administered IL-2 further decreases its effective-
ness. However, local administration of IL-2 has been more
successful and has resulted in the control of malignant
effusions and remission of established lesions (11, 13 – 15).
In this regard, a targeted accumulation of the cytokine IL-2 at
the tumor microenvironment by conjugating it to the
homodimeric scFv L19 appears to be an attractive concept to
enhance the therapeutic index of IL-2 and at the same time
diminish its toxic side effects. This study was therefore designed
to evaluate the therapeutic efficacy of the recombinant targeted
immunocytokine L19-IL-2 in pancreatic carcinoma.
Materials and Methods
Human tissue samples. Nineteen pancreatic carcinoma, 15 chronic
pancreatitis, and 11 normal pancreatic tissue samples were obtained
from individuals who underwent surgical resection at the Department
of Surgery at Charité University Hospital. This study was approved by
the local ethics committee and all patients gave written informed
consent prior to surgery.
Cell culture. The human pancreatic carcinoma cell lines MiaPaca
(American Type Culture Collection), DanG, and the murine lymphoma
cell line YAC-1 (DSMZ) were cultured as described previously (16, 17).
MiaPaca cells, stably transfected with an angiopoetin-2 DNA construct,
were maintained as MiaPaca wild-type cells, except for the addition of
hygromycin B (400 Ag/mL; Invitrogen).
Animals. Female NMRI nude mice (age, f10 weeks; weight, 21-25 g)
were purchased from Bomholtgard. Animal care followed institutional
guidelines and all experiments were approved by local animal research
authorities.
Tumor implantation and in vivo treatment. Three orthotopic
xenograft mouse models of pancreatic carcinoma were established,
including two nonmetastatic models by injection of wild-type cells of
the human pancreatic carcinoma cell lines DanG and MiaPaca and one
metastatic model by implantation of MiaPaca cells stably transfected
with an angiopoetin-2 DNA construct (MiaPaca-A2).3 Orthotopic
transplantation was carried out as described (18). In brief, a median
laparotomy was done under deep general anesthesia and the pancreas
was exposed. Aliquots of 1  106 tumor cells were injected into the
head of the pancreas. The pancreas was replaced and the abdominal
wall was closed. Therapy was started 7 days (DanG), 40 days (MiaPaca),
or 60 days (MiaPaca-A2) after tumor cell inoculation when solid
tumors and metastases (MiaPaca-A2) had formed. Groups of 8 to
12 mice were treated with vehicle (0.9% saline) or 1.43 or 4.29 MIU/kg
IL-2 equivalents as either untargeted IL-2 or L19-IL-2 (provided by
Philogen) or equimolar amounts of L19 (783 Ag/kg) for 10 days or 250
mg/kg gemcitabine (Lilly) once weekly.
3
Established byA. Scholz, unpublished data.
Clin Cancer Res 2008;14(15) August 1, 2008 4952
Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
At the end of each therapy, the tumor volume was calculated using
the formula: length  width  depth  p / 6. The area of lymph nodes
of the metastatic MiaPaca-A2 tumor model was calculated by
determining the largest diameter and its perpendicular diameter and
computing the product of the two measurements.
Depletion of natural killer cells. Natural killer (NK) cell depletion
was done as described previously (19). Briefly, mice received six
injections i.p. with 50 AL anti-asialo-GM-1 antibody (Wako Chemicals)
every fourth day, starting 3 days before tumor cell injection. The level of
NK cell depletion was monitored by flow cytometry and cytotoxicity
assay.
Preparation of spleen mononuclear cells. Spleens were removed
under deep general anesthesia and digested with collagenase
(Worthington) for 1 h at 37jC. Subsequently, contents were forced
through a 100-Am cell strainer and washed twice with HBSS.
Mononuclear cells were separated from the cell suspension via Ficoll-
Hypaque (Amersham) density gradient centrifugation.
AlamarBlue cytotoxicity assay. Cytotoxic activity of isolated
murine spleen mononuclear cells against YAC-1 cells was examined
using a 24 h AlamarBlue cytotoxicity assay (BioSource) as detailed
previously (20).
Flow cytometry. Spleen mononuclear cells (1  106) were incubated
with 1 Ag phycoerythrin-labeled anti-NK1.1 for 15 min at 4jC. After
washing with PBS, the presence of fluorescence-positive cells was
analyzed on a FACSCalibur using CellQuest software (BD).
Immunohistochemical analysis. Paraformaldehyde-fixed, 4-Am-thick
frozen sections of human and murine tissue samples were analyzed
immunohistochemically using the avidin-biotin technique. Slides were
incubated for 1 h with the following primary antibodies: biotinylated
L19-IL-2 (30 Ag/mL; provided by Bayer-Schering Pharma), monoclonal
mouse anti-human CD31 (Dako) at 1:100 dilution, rat anti-mouse
CD11b (BD Biosciences) at 1:50 dilution, biotinylated mouse anti-
mouse CD161b/c/NK1.1 (PK136) antibody (Serotec) at 1:50 dilution,
and biotinylated mouse anti-human pancytokeratin (C11) antibody
(Santa Cruz Biotechnology) at 1:50 dilution. Proliferating cells were
detected with a monoclonal mouse anti-human Ki-67 antibody
(Dianova), applying the Animal Research Kit. Apoptotic cells were
identified by terminal deoxynucleotidyl transferase – mediated dUTP
nick end labeling assay using TumorTACS In situ Apoptosis Detection
Kit (R&D Systems). Both procedures were carried out following the
manufacturer’s directions, except that 3-amino-9-ethylcarbazole was
used as the substrate chromogen. Negative controls were included by
omitting the primary antibody (Animal Research Kit) or the terminal
deoxynucleotidyl transferase (terminal deoxynucleotidyl transferase –
mediated dUTP nick end labeling). In the end, sections were counter-
stained with hematoxylin and mounted permanently. Quantitative
immunohistochemical analysis for the area of vital (nonnecrotic)
tumor tissue was carried out by computer-aided imaging analysis using
Axiovision 4.2 software (Zeiss). A red color threshold was interactively
set by the examiner for each antibody to select the red stained
immunopositive area. All pixels that met the defined threshold were
then transferred into a binary image and analyzed. For the quantifica-
tion of proliferating, apoptotic, and inflammatory cells, 20 randomly
selected measurement areas (680  510 Am; magnification, 20) of
each tumor specimen were analyzed. For evaluation of ED-B fibronectin
expression, 10 areas (1,390  1,000 Am; magnification, 10) were
scanned. The results represent the positive staining as a percentage of
measurement area (inflammatory cells and ED-B fibronectin) or
the ratio of positive tumor cells as a percentage of total tumor cells
(Ki-67 and terminal deoxynucleotidyl transferase – mediated dUTP nick
end labeling), respectively. For quantitative analysis of necrotic tumor
area, H&E sections were scanned at low power (2.5). Total and
necrotic tumor areas were marked on the screen, calculated by the
image analysis system, and the percentage of necrotic/total tumor area
was determined.
Extraction of DNA and DNA PCR. Genomic DNA of mouse lymph
nodes was extracted using the QIAamp DNA Mini Kit (Qiagen).
www.aacrjournals.org

Fig. 1. Expression of ED-B fibronectin in human pancreatic carcinoma and
orthotopic nude mouse models of pancreatic cancer. A, sections of frozen tissue
were stained with L19-IL-2 and a monoclonal CD31antibody. The antigen-antibody
complexes stain red. Representative examples of ED-B fibronectin (a, c, e, g,
and h) and CD31 (b, d, and f) immunostaining of human nontransformed pancreas
(a and b), chronic pancreatitis (c and d), and ductal adenocarcinoma (e and f)
and DanG (g) and MiaPaca (h, right margin) pancreatic cancer xenografts and
adjacent residual nontransformed pancreatic tissue (h, left margin). Bar, 100 Am.
P, pancreas; T, tumor. B, morphometric quantification of ED-B fibronectin
expression. *, P < 0.001, versus human pancreatic carcinoma.
A specific 850-bp fragment of the a-satellite region of the human
chromosome 17 was detected by PCR using the following primers
5¶-GGGATAATTTCAGCTGACTAACACG-3¶ and 5¶-TTCCGTTTAGT-
TAGGTGCAGTTATC-3¶ as described (21).
www.aacrjournals.org
Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
L19-IL2 Treatment in Pancreatic Cancer
Serum variables. Mouse blood samples were obtained by retro-
orbital bleeding. Serum CA 19-9 levels were measured by a fully
automated chemiluminescence immunoassay on an ADVIA Centaur
(Bayer) and lipase levels were determined using a kinetic colorimetric
assay on an ADVIA 2400 (Bayer).
Statistical analysis. Statistical differences were evaluated by
two-sided Mann-Whitney U test and Fisher’s exact test and correlation
was assessed by linear regression using GraphPad statistical software
(GraphPad Software). Differences were considered statistically signifi-
cant at P < 0.05.
Results
ED-B fibronectin is selectively expressed in human pancreatic
carcinoma. We studied the expression pattern of ED-B
fibronectin in human nontransformed pancreatic tissue,
chronic pancreatitis, and ductal adenocarcinoma (Fig. 1A, a,
c, and e). In parallel, serial sections from the same tissues were
stained with a monoclonal antibody against the endothelial
cell – specific antigen CD31 (Fig. 1A, b, d, and f). All but one
carcinoma sample analyzed (n = 19) revealed a specific
expression of ED-B fibronectin at the abluminal side of tumor
blood vessels and in the tumor stroma (Fig. 1A, e and f). By
contrast, nontransformed pancreatic tissue (n = 11; Fig. 1A,
a and b) and chronic pancreatitis (n = 15; Fig. 1A, c and d)
showed no ED-B fibronectin signal. Morphometric quantifica-
tion of ED-B fibronectin in pancreatic cancer recorded the
presence of ED-B fibronectin in 7.35 F 1.60% of total tissue
area compared with <0.1% in nontransformed pancreatic tissue
and chronic pancreatitis (Fig. 1B). Thus, ED-B fibronectin
provides the specific and selective overexpression required for
an ED-B fibronectin-based targeted therapy in pancreatic
cancer.
Development of an orthotopic mouse model for pancreatic
cancer. To investigate the therapeutic efficacy of an ED-B
fibronectin targeted therapy, we established orthotopic models
for pancreatic cancer by injection of human wild-type DanG or
MiaPaca pancreatic cancer cells or MiaPaca cells with stable
overexpression of angiopoetin-2 (MiaPaca-A2) into the
pancreas of nude mice. Orthotopic transplantation resulted
in extensive local tumor growth with invasion of adjacent
normal pancreatic tissue and neighboring organs. Moreover,
MiaPaca-A2 tumors showed metastatic dissemination into
intra-abdominal lymph nodes and liver. The immunohisto-
chemical analysis of the orthotopically grown pancreatic
tumors consistently revealed a strong cuff-like immunostaining
for ED-B fibronectin around tumor blood vessels and in the
stromal compartment (Fig. 1A, g and h, right). In comparison
with the human situation, the extent of ED-B fibronectin
expression was higher in DanG tumors (22.33 F 0.65%;
Fig. 1B) but equal in MiaPaca tumors (7.96 F 1.32%; Fig. 1B).
Thus, the observed tumor growth and immunohistochemical
findings compare well with the human situation and validate
the animal models as a relevant test system for therapeutic
in vivo studies with the targeted immunocytokine L19-IL2.
Targeted immunocytokine L19-IL-2 exerts significant antitumor
activity against established pancreatic cancer. To determine the
efficacy of L19-IL-2 in pancreatic cancer, mice bearing
orthotopic DanG tumors of 3.1 F 1.05 mm3 average volume
were randomized to receive vehicle or 1.43 or 4.29 MIU/kg
IL-2 equivalents of either L19-IL-2 or untargeted IL-2 for
10 days. Whereas IL-2 treatment had no effect on tumor growth
4953 Clin Cancer Res 2008;14(15) August 1, 2008
Cancer Therapy: Preclinical

Fig. 2. Therapeutic effects of L19-IL-2 on orthotopic pancreatic cancer in nude mice. A to E, nude mice bearing orthotopic DanG pancreatic tumors were randomly
treated. A, mean F SE tumor volume of each treatment group. Note that treatment with L19-IL-2 was significantly superior to treatment with equimolar amounts of IL-2.
*, P < 0.004, versus control; +, P = 0.0003, versus equimolar amounts of IL-2. B, percentage of tumor volume from vehicle-treated controls. Note that the naked L19 antibody
achieved no therapeutic effect, whereas equimolar amounts of L19-IL-2 led to a significant decrease of the primary tumor volume. *, P V 0.004, versus control and
scFv L19. C, serum CA19-9 levels from mice bearing orthotopic DanG tumors and tumor-free animals were determined and CA19-9 level versus tumor volume was plotted for
each individual animal (n = 56, r 2 = 0.54, P < 0.0001). D, CA 19-9 levels were determined at the designated time points. Points, mean of control or treatment group; bars,
SE. *, P < 0.05. E, determination of serum lipase. Mean F SE for each treatment group. F, nude mice bearing orthotopically transplanted MiaPaca pancreatic cancer were
treated as specified. Mean F SE tumor volume of each treatment group. Of note, L19-IL-2 treatment was significantly superior to treatment with equimolar amounts of IL-2.
*, P V 0.0007, versus control; +, P = 0.0003, 1.43 MIU/kg body weight L19-IL-2 versus equimolar amounts of IL-2; b, P = 0.0513, 4.29 MIU/kg body weight L19-IL-2
versus equimolar amounts of IL-2.

in the lower dose and only a minor effect in the higher dose,
administration of L19-IL-2 significantly reduced tumor volume
to 21.4% (1.43 MIU/kg) and 2.7% (4.29 MIU/kg) compared
with control mice (Fig. 2A). This result clearly shows that
L19-IL-2 is superior to the untargeted cytokine. As shown in a
separate experiment, this was not due to an IL-2-independent
antitumor activity of the scFv L19 antibody fragment, because
administration of 4.29 MIU/kg L19-IL-2 reproducibly led to a
significant reduction of tumor load, whereas equimolar
amounts of the naked scFv L19 were therapeutically inactive
(Fig. 2B).
CA 19-9 as a response marker of therapy. Orthotopic tumors
do not allow the continuous evaluation of tumor size by caliper
measurements. For this reason, we established CA 19-9 as a
valuable serum marker to monitor both tumor progression and
therapeutic effects of the targeted immunocytokine L19-IL-2.
It was possible to show a significant correlation between tumor
volume and serum concentrations of CA 19-9 (r 2 = 0.54,
P < 0.0001; Fig. 2C). Accordingly and in agreement with the
therapeutic effects on tumor burden described above, CA 19-9
levels were significantly reduced following L19-IL-2 treatment,
whereas no changes were visible in the IL-2-treated group (data
not shown). These results prompted us to evaluate the time-
response relationship of L19-IL-2-induced tumor growth
inhibition. Interestingly, a L19-IL-2-dependent tumor growth
delay in terms of 2-fold lower CA 19-9 levels when compared
with controls was observed as early as day 3 after start of
treatment. This difference continued to increase until the end of
experiment. Here, a >15-fold increase in serum CA 19-9 levels
was detectable in the control group, whereas CA 19-9 levels

Clin Cancer Res 2008;14(15) August 1, 2008 4954 www.aacrjournals.org

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
 L19-IL2 Treatment in Pancreatic Cancer

were comparable with baseline levels (Fig. 2D) in animals 4.29 MIU/kg. In addition to the anticipated reduction of
treated with L19-IL-2. primary tumor volume, L19-IL-2 treatment also decreased the
Targeted immunocytokine L19-IL-2 does not elicit an unspecific mean lymph node area to <20% of the values obtained in
inflammatory response. Having proven the therapeutic effect of vehicle-treated controls (P = 0.0120; Fig. 3B, compare a and b).
L19-IL-2, we next ensured that IL-2 activity and subsequent To determine, if changes in lymph node area were due to
inflammatory response is confined to the tumor not provoking infiltrating tumor cells, lymph nodes were immunostained for
an acute pancreatitis in adjacent nontransformed pancreatic the presence of human cytokeratin to detect epithelial cells of
tissue. For this purpose, serum lipase levels were determined as human origin. This immunostaining with a pancytokeratin
a surrogate marker of pancreatitis in mice treated with vehicle, antibody revealed large, polymorph cells with strong immu-
IL-2, or L19-IL-2. However, neither treatment increased serum noreactivity, indicative of tumor cells, in 80% of control mice
lipase concentrations (Fig. 2E). This finding was further (Fig. 3B, c) compared with 20% of L19-IL-2-treated mice. In
corroborated by the histologic analysis of pancreatic tumors contrast, lymph nodes of L19-IL-2-treated mice more frequently
and residual nontransformed pancreatic tissue of animals contained large quantities of small, granulocyte-like cells
treated with L19-IL-2 using conventional H&E staining
and immunostaining with antibodies directed against NK1.1
(NK cells) and CD11b (predominantly macrophages), respec-
tively. Of note, no signs of acute pancreatitis were detectable
and inflammatory cells selectively accumulated in the tumor,
sparing adjacent nontransformed pancreatic tissue (data not
shown). Thus, L19-IL-2 treatment did not cause pancreatitis as
an adverse effect.
Therapeutic effects of L19-IL-2 on pancreatic cancer are not cell
type specific. To investigate whether L19-IL-2-mediated reduc-
tion of tumor growth could be reproduced in other pancreatic
cancer cell lines, mice bearing orthotopic MiaPaca pancreatic
tumors were assigned to treatment with the same conditions as
specified above. Whereas untargeted IL-2 was not therapeutic at
low dosage, equimolar amounts of L19-IL-2 reduced tumor
load by 90%. At the high dosage, untargeted IL-2 achieved a
50% inhibition of tumor growth compared with an 83%
reduction of tumor volume by L19-IL-2 (Fig. 2F).
L19-IL-2 treatment inhibits pancreatic cancer lymph node
metastasis. Having confirmed the antitumor action of
L19-IL-2 on primary pancreatic tumors, we next explored its
potency in metastatic disease using the metastatic MiaPaca-A2
pancreatic tumor. Consistently, all lymph nodes infiltrated
by cancerous cells and all liver metastases showed immunore-
activity for ED-B fibronectin (Fig. 3A, a), whereas noninfiltrated
lymph nodes (Fig. 3A, b) and liver did not.
Based on this observation, mice bearing primary tumors and
metastases were randomly treated with vehicle or L19-IL-2 at

Fig. 3. Therapeutic effects of L19-IL-2 on metastatic disease in an orthotopic

mouse model of pancreatic cancer. A, lymph nodes of nude mice bearing metastatic
MiaPaca-A2 pancreatic tumors were analyzed for ED-B fibronectin expression.
Tumor-positive lymph nodes (a) showed a strong immunoreactivity for ED-B
fibronectin, whereas tumor-free lymph nodes revealed no ED-B fibronectin signal
(b). Bar, 100 Am. B to D, mice bearing metastatic MiaPaca-A2 pancreatic cancer
were treated. B, photographs show the macroscopic aspect of a representative liver
hilus of a control (a) and a L19-IL-2-treated (b) mouse (L, liver). Note that the
liver hilus lymph node (arrow) of the L19-IL-2-treated animal (b) is visibly smaller
when compared with vehicle-treated control (a). Lymph nodes were analyzed for
tumor infiltration by immunostaining for pancytokeratin. Representative examples of
a lymph node of a vehicle-treated (c) and a L19-IL-2-treated mouse (d). A strong
immunostaining of large, polymorph cells is observed in the lymph nodes of the
control mice (c). Small granulocyte-like, cytokeratin-positive cells are typically
detectable in lymph nodes of L19-IL-2-treated mice (d). Bar, 100 Am. C, lymph
nodes were analyzed by DNA-PCR for detection of a human specific 850-bp
fragment on the a-satellite DNA on human chromosome 17. Representative results
from vehicle-treated (lanes 3-7) or L19-IL-2-treated mice (lanes 8-12). A lymph
node of a tumor-free mouse and a MiaPaca primary pancreatic cancer served as
negative (lane 1) and positive (lane 2) controls, respectively. D, summarized results
of the calculated area of lymph nodes (left,Yaxis, pale columns) and the
human-specific DNA PCR (right,Yaxis, dark columns). Mean F SE lymph node area
(pale columns ; * P = 0.0120) or percentage of mice having PCR-positive lymph
nodes (dark columns ; * P = 0.0275).

www.aacrjournals.org 4955 Clin Cancer Res 2008;14(15) August 1, 2008

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
Cancer Therapy: Preclinical
Fig. 4. Long-term effects of L19-IL-2 and efficacy of treatment with L19-IL-2 and
gemcitabine. A, 10 nude mice bearing orthotopic DanG pancreatic cancer were
treated with 4.29 MIU/kg body weight L19-IL-2 and were subsequently left
untreated. CA 19-9 levels were determined at the indicated time points. B, mice
bearing orthotopic DanG pancreatic tumors were subjected to i.p. treatment with
250 mg/kg body weight gemcitabine once weekly, i.v. injection of 4.29 MIU/kg
body weight L19-IL-2 for 10 consecutive days, or vehicle treatment. Mean F SE of
each treatment group. *, P < 0.01, versus control; , P < 0.01, versus gemcitabine.
positive for cytokeratin (Fig. 3B, d). The granulocyte-like
appearance raised the possibility that these cells were detected
due to a cross-reaction of anti-cytokeratin antibodies with
granule-associated epitopes of mouse granulocytes, which has
been reported previously (22). To verify this hypothesis, lymph
nodes were screened for infiltration of human cancer cells by a
sensitive DNA-PCR-based approach. Indeed, amplificates of a
human-specific a-satellite DNA were only detected in lymph
nodes of 33% of L19-IL-2-treated mice (Fig. 3C, lanes 8-12) but
in 100% of control mice (Fig. 3C, lanes 3-7). Thus, these results
corroborate our hypothesis of a false-positive cytokeratin
immunostaining of small, granulocyte-like cells in lymph
nodes of L19-IL-2-treated animals and show that L19-IL-2
treatment had either prevented or diminished lymph node
metastases (P = 0.0275; Fig. 3D, dark columns).
L19-IL-2 treatment yields considerable long-term therapeutic
effects. To explore whether L19-IL-2 is capable of sustaining
long-term remission, mice bearing orthotopic DanG pancreatic
cancer were treated with 4.29 MIU/kg L19-IL-2. Subsequently,
mice were left untreated and tumor relapse was monitored by
biweekly measurement of serum CA 19-9 levels. Between the
Clin Cancer Res 2008;14(15) August 1, 2008
Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
fourth and the eighth weeks after termination of therapy, 5 of
10 mice experienced a marked elevation of serum tumor
marker associated with tumor relapse and subsequently died
due to tumor burden. Four of 10 mice maintained baseline
serum CA 19-9 levels throughout the entire 3 months
observation period after cessation of therapy. The autopsies
done on these animals revealed no signs of tumor relapse in
four mice, and these animals were therefore considered cured
(Fig. 4A). One animal developed a dermal tumor and died in
week 4 without any increase of serum CA 19-9 level.
L19-IL-2 treatment has superior efficacy compared with
gemcitabine. Gemcitabine is considered a standard first-line
treatment for patients with advanced pancreatic cancer. We
therefore compared L19-IL-2 and gemcitabine treatment in
mice bearing orthotopic DanG tumors. Gemcitabine reduced
pancreatic tumor volume to 28% compared with control mice.
Of note, L19-IL-2 treatment compared superior with gemcita-
bine therapy as indicated by a 2.3-fold improved tumor
reduction (Fig. 4B).
Effects of L19-IL-2 on tumor necrosis, apoptosis, and
proliferation. To identify the mechanisms underlying L19-
IL-2-induced tumor remission, we initially analyzed DanG
pancreatic tumor samples by conventional H&E staining. One
of the most intriguing features of L19-IL-2-treated tumors was
extensive tumor necrosis (Fig. 5A, b). Indeed, in mice receiving
L19-IL-2, necrosis comprised 42 F 10.86% of the tumor area,
whereas only 2.5 F 2.177% of the tumor area was necrotic in
untreated controls (P = 0.0159; Fig. 5A). The determination of
the apoptotic index in the residual vital tumor tissue via
terminal deoxynucleotidyl transferase – mediated dUTP nick
end labeling assay revealed no difference between both groups
(Fig. 5B). By contrast, L19-IL-2-treated tumors displayed a
4.6-fold reduced fraction of proliferating cells based on Ki-67
immunostaining (P = 0.0177; Fig. 5C).
Identification of the immune effector cells mediating L19-
IL-2-induced antitumor response. Apart from the extensive
necrosis, H&E staining of L19-IL-2 treated tumors showed a
round cell inflammatory infiltrate, which was absent in
untreated controls. To further characterize the immune effector
cells mediating L19-IL-2-induced tumor growth inhibition,
tumor specimen were immunohistochemically assessed for NK
cells and for CD11b+ cells, which include predominantly
macrophages. Indeed, a >4-fold increase in CD11b+ cells was
detectable in tumors of L19-IL-2-treated animals (Fig. 6A), and
even more striking, NK cells were increased by >70-fold
(Fig. 6B), suggesting that NK cells represent major players in
the L19-IL-2-induced antitumor response in this animal model.
To further substantiate this suspected role, we tested the effect
of NK cell depletion via an anti-asialo-GM-1 antibody on the
therapeutic efficacy of L19-IL-2. The NK depletion abrogated
the effects of L19-IL-2 treatment such that tumor volume did
not significantly differ from untreated controls (Fig. 6C). From
these results, we conclude that NK cell – mediated mechanisms
are required for the therapeutic efficacy of the targeted
immunocytokine L19-IL-2.
Discussion
IL-2 is an approved drug for the treatment of metastatic renal
cell carcinoma and melanoma. To specifically increase the
concentration of IL-2 at the tumor site, locoregional treatment
4956 www.aacrjournals.org
 L19-IL2 Treatment in Pancreatic Cancer

schedules were designed and investigated. (11, 12). High
concentrations of IL-2 are indeed essential for a therapeutic
effect and have been exemplified by the application of IL-2 in
patients with advanced pancreatic cancer via arterial or portal
venous catheters in combination with polychemotherapy
(23, 24). Unfortunately, all locoregional applications are
limited to tumors that are accessible from the outside without
having the possibility to deliver the cytokine to distant
metastases. This hurdle can be overcome by a highly selective
compound such as the targeted immunocytokine L19-IL-2. The
most important prerequisite for a novel antibody-targeted
delivery therapy like L19-IL-2 is the tumor-specific expression
and accessibility of a target molecule within the body. ED-B
fibronectin fulfills these requirements because (a) this molecule
is specifically up-regulated during tumor angiogenesis and
tumor growth (6, 25, 26) and (b) it is accessible for antibody-
targeted therapies. This has been shown with the ED-B-specific
scFv L19, which was successfully used for tumor imaging (10)
and the delivery of various effector molecules in animal studies
(26 – 29).
This is the first study that shows a specific overexpression
of ED-B fibronectin in human pancreatic carcinoma, whereas
no ED-B fibronectin was detectable in normal pancreas or
chronic pancreatitis. Our immunohistochemical data strongly

Fig. 5. Effects of L19-IL-2 on tumor necrosis, apoptosis, and proliferation. DanG tumors from mice that had received either vehicle or 4.29 MIU/kg body weight L19-IL-2
were examined by conventional H&E staining (A, a and b), terminal deoxynucleotidyl transferase ^ mediated dUTP nick end labeling assay (B, a and b), or Ki-67
staining (C, a and b). Right, quantification of necrosis (A ; *, P = 0.0159), apoptosis (B), and proliferation (C ; *, P = 0.0177). Mean F SE percentages of each group. Bar,
500 Am (A) and 50 Am (B and C).

www.aacrjournals.org 4957 Clin Cancer Res 2008;14(15) August 1, 2008

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
Cancer Therapy: Preclinical

Fig. 6. Identification of the immune effector cells mediating L19-IL-2-induced tumor regression. A and B, DanG tumors from mice that had received either vehicle or
L19-IL-2 as indicated were examined with a CD11b antibody (A) and a NK1.1antibody (B). Left, representative examples of vehicle-treated (A and B, a) and L19-IL-2-treated
mice (A and B, b). Bar, 50 Am. Right, quantification of infiltrating CD11b+ cells (A) and NK cells (B). *, P = 0.0159 (A) and 0.0012 (B). C, nude mice bearing
DanG orthotopic tumors were randomly assigned to the indicated treatment groups. Columns, mean of tumor volumes of each group; bars, SE. *, P < 0.01, versus any
other group.

supported the evaluation of the therapeutic efficacy of the
targeted immunocytokine L19-IL-2 in mouse models for this
devastating disease. There is increasing evidence that the ectopic
or orthotopic environment has a different effect on tumor cell
protein expression, tumor growth, invasiveness, angiogenesis,
metastasis, drug delivery, and sensitivity to therapeutic agents
(30). For this reason, we developed orthotopic mouse models
for pancreatic cancer, which accurately recapitulate the tumor
behavior and the clinical course of this deadly disease.
In both models, we strongly showed an ED-B fibronectin
expression pattern, which is comparable with the human
situation in both primary tumors and lymph node and liver
metastases. Here, we report the first study that addresses the
therapeutic efficacy of L19-IL-2 in a tumor grown in its natural
microenvironment. In this orthotopic setting, we could clearly
show that the systemic i.v. administration of clinically relevant
doses of L19-IL-2 efficiently inhibited the growth of established
primary pancreatic tumors and distant metastases, whereas
equivalent concentrations of untargeted IL-2 had no or only
minor therapeutic effects. It is noteworthy that the therapeutic
effect of L19-IL-2 could be shown in two independent mouse
models with different quantitative level of ED-B fibronectin
expression (DanG, 22.33%; MiaPaca, 7.96%). This observation
was unexpected, yet the density of ED-B fibronectin visualized

Clin Cancer Res 2008;14(15) August 1, 2008 4958 www.aacrjournals.org

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
 L19-IL2 Treatment in Pancreatic Cancer

by immunohistochemistry must not necessarily correlate
with the in vivo accumulation of the targeted immunocytokine
L19-IL-2 in the tumor tissue. It may very well be that, in
comparison with DanG, the apparently low target density in the
MiaPaca tumor model can be compensated by a more
functional vascular network leading to a higher accumulation
of L19-IL-2 at the tumor site. Different levels of L19-IL-2
accumulation could also explain the induction of a complete
and permanent remission in 40% of the animals, whereas
untargeted IL-2 showed tumor regrowth after treatment. If our
hypothesis is true, the stratification of patients who will most
likely benefit from a targeted immunocytokine therapy will be
possible using radioactively labeled L19. Future experiments in
this direction are currently being done.
In this context, it is also important to note that, until today,
the mechanisms underlying the antitumor activity of IL-2 are
still not fully understood. In our study, one of the most
intriguing features of L19-IL-2-treated tumors was an extensive
tumor necrosis, which was most prominent in the central part
of the tumor. In this region, few leukocytes were detectable.
Therefore, tumor cell death caused by direct cell-to-cell contact
with NK cells, macrophages, or other immune effector cells is
unlikely. Indeed, recent evidence supports the occurrence of a
local vascular leak syndrome provoked by lymphokines and
nitric oxide produced by IL-2-activated immune effector cells
(31, 32). Furthermore, direct cytotoxic effects of IL-2 (12) and
activated NK cells (33, 34) on the endothelium were reported.
Considering that (a) endothelial cells seem to migrate along
extracellular matrix structures containing ED-B fibronectin
(35) and (b) L19-IL-2 is selectively accumulating within the
ED-B fibronectin-rich proangiogenic tumor microenviron-
ment, we postulate that L19-IL-2-triggered direct and/or
indirect mechanisms have a detrimental effect on both the
tumor vasculature and the tumor cells themselves, which are
per se insensitive to IL-2-treatment (36). Therefore, the
considerable decrease of Ki-67-positive tumor cells in the
tumor of L19-IL-2-treated animals can only be attributed to
cytokines such as IFN-g and tumor necrosis factor-a, which
are released by the large amounts of tumor-infiltrating NK
cells and macrophages. Both IFN-g and tumor necrosis factor-
a are known to directly inhibit the growth of pancreatic
cancer cells (37, 38).
We could bolster this hypothesis by showing a L19-IL-
2-triggered 4-fold increase of tumor-infiltrating macrophages
and a >70-fold increase of NK cells. These data suggest that
macrophages were not the main drivers for the therapeutic
efficacy of L19-IL-2 in our study. In line with this notion, it has
been established previously that depletion of macrophages
could only attenuate but not abrogate the antitumor activity of
IL-2, indicating the contribution of additional mechanisms
(39). In our study, we could pinpoint the NK cells as the
driving force, which on depletion completely abolished the
therapeutic efficacy of L19-IL-2.
Our findings are in perfect agreement with published data,
which report on (a) promising antitumor effects in addition to
extravasation and infiltration of tumor tissues by NK cells in
vitro and in vivo (40) and (b) NK cells that were identified as the
primary target cell population for IL-2 in preclinical and clinical
settings (41).
In conclusion, our preclinical data strongly support the
initiation of clinical studies using the targeted immunocytokine
L19-IL-2 in pancreatic cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.

References
1. Parker SL, Tong T, Bolden S, et al. Cancer statistics,
1997. CA Cancer J Clin 1997;47:5 ^ 27.
2. Rosewicz S, Wiedenmann B. Pancreatic carcinoma.
Lancet 1997;349:485 ^ 9.
3. Cohen SJ, Pinover WH, Watson JC, et al. Pancreatic
cancer. CurrTreat Options Oncol 2000;1:375 ^ 86.
4. MacKenzie MJ. Molecular therapy in pancreatic ade-
nocarcinoma. Lancet Oncol 2004;5:541 ^ 9.
5. Schneider G, Siveke JT, Eckel F, et al. Pancreatic can-
cer : basic and clinical aspects. Gastroenterology
2005;128:1606 ^ 25.
6. Alessi P, Ebbinghaus C, Neri D. Molecular targeting
of angiogenesis. Biochim Biophys Acta 2004;1654:
39 ^ 49.
7. Reddy LH. Drug delivery to tumours: recent strate-
gies. J Pharm Pharmacol 2005;57:1231 ^ 42.
8. Halin C, Zardi L, Neri D. Antibody-based targeting of
angiogenesis. News Physiol Sci 2001;16:191 ^ 4.
9. Viti F, Tarli L, Giovannoni L, et al. Increased binding
affinity and valence of recombinant antibody frag-
ments lead to improved targeting of tumoral angio-
genesis. Cancer Res 1999;59:347 ^ 52.
10. Santimaria M, Moscatelli G,Viale GL, et al. Immuno-
scintigraphic detection of the ED-B domain of fibro-
nectin, a marker of angiogenesis, in patients with
cancer. Clin Cancer Res 2003;9:571 ^ 9.
11. Bubenik J, Den Otter W, Huland E. Local cytokine
therapy of cancer: interleukin-2, interferons and relat-
ed cytokines. Cancer Immunol Immunother 2000;49:
116 ^ 22.
12. Baluna R, Rizo J, Gordon BE, et al. Evidence for a
structural motif in toxins and interleukin-2 that may
be responsible for binding to endothelial cells and ini-
tiating vascular leak syndrome. Proc Natl Acad Sci
U S A 1999;96:3957 ^ 62.
13. Den OtterW, Dobrowolski Z, Bugajski A, et al. Intra-
vesical interleukin-2 in T1 papillary bladder carcinoma:
regression of marker lesion in 8 of 10 patients. J Urol
1998;159:1183 ^ 6.
14. Baselmans AH, Koten JW, Battermann JJ, et al. The
mechanism of regression of solid SL2 lymphosarcoma
after local IL-2 therapy. Cancer Immunol Immunother
2002;51:492 ^ 8.
15. Krastev Z, Koltchakov V, Popov D, et al. A case of
hepatocellular carcinoma (HCC): treatment with local
application of alcohol and interleukin 2 (IL-2). Hepato-
gastroenterology 2003;50:1647 ^ 9.
16. Lode HN, Xiang R, Dreier T, et al. Natural killer cell-
mediated eradication of neuroblastoma metastases to
bone marrow by targeted interleukin-2 therapy. Blood
1998;91:1706 ^ 15.
17. PlathT, Peters M, Detjen K, et al. Overexpression of
pRB in human pancreatic carcinoma cells: function in
chemotherapy-induced apoptosis. J Natl Cancer Inst
2002;94:129 ^ 42.
18. Alves F, Contag S, Missbach M, et al. An orthotopic
model of ductal adenocarcinoma of the pancreas in
severe combined immunodeficient mice representing
all steps of the metastatic cascade. Pancreas 2001;
23:227 ^ 35.
19. Peron JM, Couderc B, Rochaix P, et al. Treatment of
murine hepatocellular carcinoma using genetically
modified cells to express interleukin-12. J Gastro-
enterol Hepatol 2004;19:388 ^ 96.
20. Nociari MM, Shalev A, Benias P, et al. A novel one-
step, highly sensitive fluorometric assay to evaluate
cell-mediated cytotoxicity. J Immunol Methods 1998;
213:157 ^ 67.
21. Becker M, Nitsche A, Neumann C, et al. Sensitive
PCR method for the detection and real-time quantifi-
cation of human cells in xenotransplantation systems.
Br J Cancer 2002;87:1328 ^ 35.
22. Streicher J, Fabian B, Herkner K, et al. Anticytoker-
atins are a potential source of false-positive indirect
immunofluorescence assays for C-ANCA. J Clin Lab
Anal 1998;12:54 ^ 9.
23. Abdel-Wahab M, El Shennawy F, Agha S, et al.
Evaluation of cell mediated immunity in advanced
pancreatic carcinoma before and after treatment with
interleukin-2 (IL-2). Hepatogastroenterology 1999;46
Suppl 1:1293 ^ 6.
24. Lygidakis NJ,Vlachos L, Raptis S, et al. Consecutive
re-explorations for final resection of initially unresect-
able pancreatic head carcinoma. Hepatogastroenter-
ology 1999;46:2229 ^ 39.
25. Castellani P,Viale G, DorcarattoA, et al.The fibronec-
tin isoform containing the ED-B oncofetal domain: a
marker of angiogenesis. Int JCancer1994;59:612 ^ 8.
26. Halin C, Rondini S, Nilsson F, et al. Enhancement of
the antitumor activity of interleukin-12 by targeted de-
livery to neovasculature. Nat Biotechnol 2002;20:
264 ^ 9.
27. Carnemolla B, Borsi L, Balza E, et al. Enhancement
of the antitumor properties of interleukin-2 by its tar-
geted delivery to the tumor blood vessel extracellular
matrix. Blood 2002;99:1659 ^ 65.

www.aacrjournals.org 4959 Clin Cancer Res 2008;14(15) August 1, 2008

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
Cancer Therapy: Preclinical
28. Borsi L, Balza E, Carnemolla B, et al. Selective tar-
geted delivery of TNFa to tumor blood vessels. Blood
2003;102:4384 ^ 92.
29. Nilsson F, Kosmehl H, Zardi L, et al. Targeted deliv-
ery of tissue factor to the ED-B domain of fibronectin,
a marker of angiogenesis, mediates the infarction of
solid tumors in mice. Cancer Res 2001;61:711 ^ 6.
30. Killion JJ, Radinsky R, Fidler IJ. Orthotopic models
are necessary to predict therapy of transplantable
tumors in mice. Cancer Metastasis Rev 1998;17:
279 ^ 84.
31. De Mik HJ, Koten JW, Maas RA, et al. Tumour re-
gression by IL-2 mediated stagnation of blood flow.
In Vivo 1991;5:679 ^ 84.
32. Sakkoula E, Pipili-Synetos E, Maragoudakis ME.
Involvement of nitric oxide in the inhibition of an-
giogenesis by interleukin-2. Br J Pharmacol 1997;
122:793 ^ 5.
Clin Cancer Res 2008;14(15) August 1, 2008
Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.
33. Albertsson PA, Basse PH, Hokland M, et al. NK cells
and the tumour microenvironment: implications for
NK-cell function and anti-tumour activity. Trends
Immunol 2003;24:603 ^ 9.
34. Di Carlo E, Meazza R, Basso S, et al. Dissimilar anti-
tumour reactions induced by tumour cells engineered
with the interleukin-2 or interleukin-15 gene in nude
mice. J Pathol 2000;191:193 ^ 201.
35. Tarli L, Balza E, Viti F, et al. A high-affinity human
antibody that targets tumoral blood vessels. Blood
1999;94:192 ^ 8.
36. Rosenberg SA. Progress in the development of
immunotherapy for the treatment of patients with
cancer. J Intern Med 2001;250:462 ^ 75.
37. Detjen KM, Farwig K, Welzel M, et al. Interferon g
inhibits growth of human pancreatic carcinoma cells
via caspase-1 dependent induction of apoptosis. Gut
2001;49:251 ^ 62.
4960
38. WatanabeT, Fuchimoto S, Matsubara N, et al. Anti-
proliferative effect on human pancreatic cancer cells of
natural human tumour necrosis factor-h combined
with natural human interferon-a or interferon-g. J Int
Med Res 1992;20:112 ^ 20.
39. Masztalerz A, Van Rooijen N, Den Otter W, et al.
Mechanisms of macrophage cytotoxicity in IL-2 and
IL-12 mediated tumour regression. Cancer Immunol
Immunother 2003;52:235 ^ 42.
40. Basse PH, Whiteside TL, Herberman RB. Use of
activated natural killer cells for tumor immunother-
apy in mouse and human. Methods Mol Biol
2000;121:81 ^ 94.
41. Janssen RA, Sleijfer DT, Heijn AA, et al. Peripheral
blood lymphocyte number and phenotype prior to
therapy correlate with response in subcutaneously ap-
plied rIL-2 therapy of renal cell carcinoma. Br J Cancer
1992;66:1177 ^ 9.
www.aacrjournals.org
The Targeted Immunocytokine L19-IL2 Efficiently Inhibits the
Growth of Orthotopic Pancreatic Cancer
Karola Wagner, Petra Schulz, Arne Scholz, et al.

Clin Cancer Res 2008;14:4951-4960.

Updated version Access the most recent version of this article at:
http://clincancerres.aacrjournals.org/content/14/15/4951

Cited Articles This article cites by 41 articles, 12 of which you can access for free at:
http://clincancerres.aacrjournals.org/content/14/15/4951.full.html#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at:
http://clincancerres.aacrjournals.org/content/14/15/4951.full.html#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Subscriptions Department at pubs@aacr.org.

Permissions To request permission to re-use all or part of this article, contact the AACR Publications
Department at permissions@aacr.org.

Downloaded from clincancerres.aacrjournals.org on April 26, 2015. © 2008 American Association for Cancer
Research.